Cite this article
1
Download
46
Views
Journal Browser
Volume | Year
Issue
Search
News and Announcements
View All
ARTICLE

Application of sparse dictionary learning to seismic data reconstruction

HAMID REZA KHATAMI1 MOHAMMAD ALI RIAHI2 MOHAMMAD MAHDI ABEDI3
Show Less
1 Petroleum, Mining and Materials Engineering, Islamic Azad University, Central Tehran Branch, Tehran, Iran.,
2 Institute of Geophysics, University of Tehran, Tehran, Iran.,
3 BCAM - Basque Center for Applied Mathematics, Bilbao, Spain.,
JSE 2023, 32(2), 185–204;
Submitted: 11 January 2023 | Accepted: 15 March 2023 | Published: 1 April 2023
© 2023 by the Authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution -Noncommercial 4.0 International License (CC-by the license) ( https://creativecommons.org/licenses/by-nc/4.0/ )
Abstract

Khatami, H.R., Riahi, M.A. and Abedi, M.M., 2023. Application of sparse dictionary learning to seismic data reconstruction. Journal of Seismic Exploration, 32:185-204. According to the principle of compressed sensing (CS), under-sampled seismic data can be interpolated when the data becomes sparse in a transform domain. To sparsify the data, dictionary learning presents a data-driven approach trained to be optimized for each target dataset. This study presents an interpolation method for seismic data in which dictionary learning is employed to improve the sparsity of data representation using improved Kth Singular Value Decomposition (K-SVD). In this way, the transformation will be highly compatible with the input data, and the data in the converted domain will be sparse. In addition, the sampling matrix is produced with the restricted isometry property (RIP). To reduce the sensitivity of the minimizer term to the outliers, we use the smooth L1 minimizer as a regularization term in the regularized orthogonal matching pursuit (ROMP). We apply the proposed method to both synthetic and real seismic data. The results show that it can successfully reconstruct seismic data.

Keywords
compressed sensing
dictionary learning
optimization
reconstruction
sparsity
References
  1. Baraniuk, R.G. and Steeghs, P., 2017. Compressive sensing: A new approach to seismicdata acquisition. The Leading Edge, 36: 642-645.
  2. Boyd, S. and Vandenberghe, ]., 2004. Convex Optimization. Cambridge University Press,New York.
  3. Candés, E.J., Romberg, J. and Tao, T., 2006. Robust uncertainty principles: Exact signalreconstruction from highly incomplete frequency information. IEEE Transact.Info.. Theory, 52: 489-509.
  4. Candés, E.J., Wakin, M.B. and Boyd, S.P., 2008. Enhancing sparsity by reweighted 1 1minimiz.. J. Fourier Analys. Applicat., 14: 877-905.
  5. Chen, Y., Ma, J. and Fomel, S., 2016. Double-sparsity dictionary for seismic noiseattenuation. Geophysics, 81, 2: V103—V116.
  6. Chen, Y., 2017. Fast dictionary learning for noise attenuation of multidimensionalseismic data. Geophys. J. Internat., 209: 21-31.
  7. Donoho, D.L., 2006. Compressed sensing. IEEE Transact. Informat. Theory, 52:1289-
  8. Donoho, D.L., Elad, M. and Temlyakov, V.N., 2005. Stable recovery of sparseovercomplete representations in the presence of noise. IEEE Transact. Informat.Theory, 52: 6-18.
  9. Duarte, M.F. and Eldar, Y.C., 2011. Structured compressed sensing: From theory toapplications. IEEE Transact. Signal Process., 59: 4053-4085.
  10. Elad, M. and Aharon, M., 2006. Image denoising via sparse and redundantrepresentations over learned dictionaries. IEEE Transact. Image Process., 15: 3736-
  11. Gilbert, A.C., Muthukrishnan, S. and Strauss, M., 2005. Improved time bounds for near-optimal sparse Fourier representations. In: Wavelets XI, 5914: 398-412. SPIE.doi: 10.1117/12.615931
  12. Herrmann, F.J. and Hennenfent, G., 2008. Non-parametric seismic data recovery withcurvelet frames. Geophys. J. Internat., 173: 233-248.
  13. Herrmann, F.J., Wason, H. and Lin, T., 2011. Compressive sensing in seismicexploration: an outlook on a new paradigm. CSEG Recorder, 36: 19-33.
  14. Iwen, M.A., 2007. A deterministic sub-linear time sparse Fourier algorithm via non-adaptive compressed sensing methods. arXiv: 0708.1211v1.doi.org/10.48550/arXiv.0708.1211
  15. Jahanjooy, S., Nikrouz, R. and Mohammed, N., 2016. A faster method to reconstructseismic data using the anti-leakage Fourier transform. J. Geophys. Engineer., 13:86-95.
  16. Kaur, H., Pham, N. and Fomel, S., 2019. Seismic data interpolation using CycleGAN.
  17. Expanded Abstr., 89th Ann. Internat. SEG Mtg., San Antonio: 2202-2206.
  18. Lan, N.Y., Zhang, F. C. and Yin, X.Y., 2022. Seismic data reconstruction based on lowdimensional manifold model. Petrol. Sci., 19: 518-533.
  19. Lotfi, M. and Vidyasagar, M., 2018. A fast noniterative algorithm for compressivesensing using binary measurement matrices. IEEE Transact. Signal Process., 66:4079-4089.
  20. Mallat, S.G. and Zhang, Z., 1993. Matching pursuits with time-frequency dictionaries.IEEE Transact. Signal Process., 41: 3397-3415.
  21. Meng, F., Yang, X., Zhou, C. and Li, Z., 2017. A sparse dictionary learning-basedadaptive patch inpainting method for thick cloud removal from high-spatialresolution remote sensing imagery. Sensors, 17: 2130.
  22. Nazari Siahsar, M.A., Gholtashi, S., Abolghasemi, V. and Chen, Y., 2017a.
  23. Simultaneous denoising and interpolation of 2D seismic data using data-driven non-negative dictionary learning. Signal Process., 141: 309-321.
  24. Nazari Siahsar, M.A., Gholtashi, S., Roshandel Kahoo, A., Chen, W. and Chen, Y.,2017b. Data-driven multitasks sparse dictionary learning for noise attenuation of 3Dseismic data. Geophysics, 82, 6: V385—V396.
  25. Needell, D. and Tropp, J.A., 2009. CoSaMP: Iterative signal recovery from incompleteand inaccurate samples. Appl. Computat. Harmon. Analis., 26: 301-321.
  26. Needell, D. and Vershynin, R., 2010. Signal recovery from incomplete and inaccuratemeasurements via regularized orthogonal matching pursuit. IEEE J. Select. TopicsSignal Process., 4: 310-316.
  27. Oguz, I., Zhang, L., Abramoff, M.D. and Sonka, M., 2016. Optimal retinal cystsegmentation from OCT images. In: Medical Imag. 2016: Image Process.. 9784:375-381. Comput. Sci. SPIE Medical Imaging.
  28. She, B., Wang, Y., Liu, Z., Cai, H., Liu, W. and Hu. G.M., 2019, Seismic impedanceinversion using dictionary learning-based sparse representation and nonlocalsimilarity. Interpretation, 7: SE51-SE67.
  29. Shi, R., Bing Li, B. and Zhang, J., 2018. Modulated signal denoising algorithm based onimproved K-SVD. IOP Conf. Series: Materials Science and Engineering 452: 1-6.doi: 10.1088/1757-899X/452/3/032064.
  30. Strohmer, T. and Heath, R.W., 2003. Grassmannian frame with applications to codingand communication. Appl. Computat. Harmon. Analys., 14: 257-275.
  31. Sun, H.M., Jia, R.S., Zhang, X.L., Peng, Y. J. and Lu. X.M., 2019. Reconstruction ofmissing seismic traces based on sparse dictionary learning and the optimization ofmeasurement matrices. J. Petrol. Sci. Engineer., 175: 719-727.
  32. Tropp, JA. and Gilbert, A.C., 2007. Signal recovery from random measurements viaorthogonal matching pursuit. IEEE Transact. Informat. Theory, 53: 4655-4666.
  33. Wang, H., Chen, W., Zhang, Q., Liu, X., Zu, S. and Chen, Y., 2020. Fast dictionarylearning for high-dimensional seismic reconstruction. IEEE Transact. Geosci.Remote Sens., 59: 7098-7108.
  34. Yu, S., Ma, J., Zhang, X. and Sacchi, M.D., 2015. Interpolation and denoising of high-dimensional seismic data by learning a tight frame. Geophysics, 80(2): V119-V132.
  35. Zhou, Y., Gao, J., Chen, W. and Frossard, P., 2016. Simultaneous source separation viapatchwise sparse representation. IEEE Transact. Geosci. Remote Sens., 54: 5271-
  36. Zu, S., Zhou, H., Wu, R., Mao, W. and Chen, Y., 2018. Hybrid-sparsity constraineddictionary learning for iterative deblending of extremely noisy simultaneous-sourcedata. IEEE Transact. Geosci. Remote Sens., 57: 2249-2262.
Share
Back to top
Journal of Seismic Exploration, Electronic ISSN: 0963-0651 Print ISSN: 0963-0651, Published by AccScience Publishing